The Identification of Potential Therapeutic Targets for Lymphoma through Mendelian Randomization

Introduction: Lymphoma belongs to an extremely heterogeneous group of diseases, and the direct regulation of protein activity has been major focus of traditional drug discovery. Further evidence is urgently needed to identify novel therapeutic targets for lymphoma treatment. Mendelian randomization (MR) is a genetic epidemiology method that uses genetic variants as predictors to estimate the causal effects of modifiable exposure on outcome. Herein, we aimed to explore the causal relationship between plasma protein and lymphoma in European individuals using MR analysis.

Methods: A two-sample mendelian randomization (TSMR) approach was employed in this study, utilizing summary statistics from the deCODE study (N=355,559) and the Fenland study (N=10,708). The primary method used for the MR analysis was the inverse variance weighted (IVW) method and wald ratio. Several tests, including Cochran's Q test and Egger's regression intercept, were conducted to identify potential heterogeneity and pleiotropy. Additionally, to further assess drug targets, protein-protein interaction analysis (PPI), enrichment analysis and functional annotation, single-cell sequencing analysis, drug prediction analysis, molecular docking analysis, and mouse genome informatics (MGI) were performed.

Results: Using MR analysis, 113 proteins associated with lymphoma risk were identified in the deCODE database. Among these, 52 proteins were verified to be relevant to lymphoma in the Fenland database. Further sensitivity analysis was performed to identify proteins exhibiting pleiotropic effects and heterogeneity. Ultimately, a total of 37 plasma proteins were found to have a causal association with lymphoma, while these proteins were robust to various MR assumptions.

To evaluate the feasibility of these 37 proteins as potential drug targets for lymphoma, we conducted a comprehensive analysis of known therapeutic drugs (n=139) for lymphoma in DrugBank, and String website was used to construct a PPI network, showing the interactions between these 37 proteins and known drug targets. In the PPI network analysis, 8 proteins, including APOBEC3G, TNFAIP6, LY75, MAPK3, SERPING1, CSF2RB, FCRL3 and LY9, were found to interact with known drug targets for lymphoma, which were considered as first-level targets.

To obtain a better understanding of the biological significance of the preceding identified proteins, enrichment analysis and tissue expression analysis were performed. These 8 proteins, the first-level targets, were significantly enriched in the organism's response to stimuli and the proliferation of leukocytes. Furthermore, the genes encoding these 8 proteins exhibited elevated expression levels in spleen, whole blood, and lymphocytes compared to the remaining 29 proteins. Then, tissue enrichment analysis showed a significant and pronounced enrichment of these 8 proteins in lymphocytes and small intestine. In PheWAS analysis, LY75, SERPING1, and CSF2RB were found to be associated with hematologic abnormalities and immune disorders, indicating that they might have a pleiotropic effect on MR analysis results. Moreover, single cell analysis revealed the enriched expression of FCRL3 and LY9 in malignant B cells.

Subsequently, we further investigated the feasibility of converting potential targets for drug discovery in lymphoma. Given that FCRL3 and LY9 were identified as crucial targets for lymphoma development, the DSigDB database was used to predict potential anti-tumor drugs targeting FCRL3 and LY9. The top 10 drugs were then selected for structural analysis by molecular docking, which revealed the highest binding affinity of FCRL3 to amiodarone, and LY9 to chrysin, suggesting the most promising drug candidates for lymphoma treatment. Moreover, MGI analysis found the association of LY75, MAPK3, CSF2RB, and LY9 with abnormalities in the hematopoietic system. Notably, the knock-out of LY9 could lead to decreased proliferation of T cell and abnormal morphology of T-helper 2 cell.

Conclusions: Our integrative analysis revealed a strong relationship between 8 plasma proteins and risk of lymphoma, and the proteins FCRL3 and LY9 showed potential as attractive targets for drug development in lymphoma. Additional research is necessary to investigate the regulatory roles and druggable potential of these candidate proteins in lymphoma development.

No relevant conflicts of interest to declare.